Actionable AI Insights & Foresights into ED Crowding
Research type
Research Study
Full title
A Study into the Actionable AI-derived Insights & Foresights for Use in Reducing Emergency Department Crowding
IRAS ID
315484
Contact name
Adrian Boyle
Contact email
Sponsor organisation
Cambridge University Hospitals
Clinicaltrials.gov Identifier
315484, IRAS
Duration of Study in the UK
3 years, 0 months, 0 days
Research summary
Emergency department (ED) overcrowding is an extremely complex, multifactorial and heterogeneous problem that has been proven to negatively impact patient safety and outcomes. In England in 2019/20, 24.8 million ED patients were treated at a cost of £3.2 billion to the NHS, and only 84% of these patients were treated within 4 hours. This figure is predicted to drop to 79% by 2025/26, meaning that supply is increasingly struggling to meet demand for emergency medicine. The aim of this study is to access historical data from all patients registered at Addenbrooke's Hospital Emergency Department between 01/01/2016 and 31/12/2021, and explore whether the application of AI/machine learning algorithms can provide ED and hospital administrative staff with actionable insights and predictions that can be used to combat causes of crowding, improve patient flow, improve patient experience and outcomes and, in doing so, increase capacity while reducing overall costs of emergency care to the NHS. The potential benefits will be the creation of better business intelligence software that is able to accurately predict key elements of demand on EDs and emergency admissions, leading to better resourcing decisions and contributing towards one of the five key aims outlined in the NHS Long Term Plan; to reduce pressure on emergency hospital services. As such, the sponsor Cambridge University Hospitals (CUH) is collaborating with electronRx, a deep tech company of data scientists and machine learning engineers who have been identified by CUH as best placed to conduct this analysis and realise the potential benefits of the study. Due to the need for a solution which can accommodate the unique challenges facing each individual ED, to ensure scalability beyond CUH to other NHS Trusts, CUH has identified AI/machine learning and their single, fully integrated EHR as having the strong potential to positively impact efforts to solve this problem.
REC name
London - Dulwich Research Ethics Committee
REC reference
22/LO/0640
Date of REC Opinion
14 Dec 2022
REC opinion
Further Information Favourable Opinion